Simulation method for finite element of laser welding temperature field based on BP neural network and genetic algorithm GA

Abstract

The invention discloses a simulation method for a finite element of a laser welding temperature field based on a BP neural network and a genetic algorithm GA, which includes the following steps. Step 1 is a simulation of the finite element of the laser welding in which a heat source model with a surface body combination consisting of a gauss surface heat source and a cylinder heat source is selected, a thermal distribution coefficient of a surface heat source, an effective radius of the surface heat source, and an effective thermal power coefficient are selected to be design variables. Moreover, an orthogonal experiment table is designed and simulation errors of the finite element of the heat source model with a surface body combination under different parameters are calculated. Step 2, the BP neural network is established, trained and tested. Step 3, the solution of a optimum parameter is found and the feasibility of optimization results is determined. Then the BP neural network is re-established and the optimization solution of the genetic algorithm is found again when there are big errors. The simulation method for the finite element of the laser welding temperature field based on the BP neural network and the genetic algorithm GA achieves the improved efficiency and precision of the traditional welding temperature field, thereby being easy to find the optimal solution of the simulation.

Description

technical field [0001] The invention relates to a laser welding finite element simulation method, in particular to a laser welding temperature field finite element simulation method based on BP neural network and genetic algorithm GA. Background technique [0002] Laser welding has been widely used in the joining process due to its advantages of high energy density, small heat-affected zone, small deformation, and strong welding flexibility. In the automobile industry, the use of tailor-made welded blanks is an important means to realize the lightweight of automobiles. It can not only meet the requirements of various parts of the body structure on material, thickness, strength and corrosion resistance, but also improve the assembly accuracy of the body and improve the Rigidity, reduce the number of parts, improve the degree of body integration. With the emphasis on welding quality and welding production efficiency, finite element simulation is widely used to reproduce the w...

AI Technical Summary

Problems solved by technology

[0004] In order to solve the technical problems of the low efficiency of the traditional simulation method and the difficulty of obtaining the optimal solution for the heat source parameter value, the present invention provides a finite element simulation method of the laser welding temperature field based on BP neural network and genetic algorithm GA, which can improve the traditional welding temperature Efficiency in field simulation, easy to get the optimal solution for simulation

Images